# AI Security: Maybelyn Plecic on Shadow AI, Non Human Identities, and AI Defense

An episode of [AI Security, AI Cloud Ops & Infrastructure | ClearTech Loop](https://invisiblenetwork.ai/podcast/1823980617), 2 Jun 2026, 19 min.

Audio: https://www.buzzsprout.com/2248577/episodes/19278105-ai-security-maybelyn-plecic-on-shadow-ai-non-human-identities-and-ai-defense.mp3

Your AI policy does not matter much if no one understands how to follow it. In this episode of ClearTech Loop, Jo Peterson talks with Maybelyn Plecic, Manager of Training and Adoption at Network to Code, about shadow AI, non human identities, and what AI defense actually means when people are already using AI to get work done. Maybelyn brings a security, compliance, training, and adoption lens to the conversation. She is CISSP certified, AWS certified, and has spent her career helping teams strengthen security posture, drive compliance initiatives, and make technical change usable. Why This Matters AI adoption is already happening inside organizations. The challenge is that governance, policy, training, and approved tools are not always keeping pace. That creates risk, but not always because people are acting recklessly. In many cases, employees are trying to move faster, automate boring work, and solve problems the official process has not solved yet. Maybelyn frames shadow AI as an IT issue, a security issue, and a trust issue. Her point is clear: if leaders want people to use AI safely, they have to make the safe path understandable, practical, and easier than the workaround. What You’ll Hear in This Episode • Why shadow AI starts with trust, not blame • How protected proof of concept environments and AI sandboxes can reduce risk • Why shared language matters when AI systems, agents, and workflows touch data • How prompt injection, AI training defaults, and history tracking create new security concerns • Why AI defense is not just a tooling conversation • How leaders can create AI guidance that teams will actually follow Key Insight AI security is not only about tools and controls. It is about whether people understand the rules, whether the approved process works, and whether organizations are willing to meet teams where the work actually happens. As Maybelyn says in the episode: “how do you expect someone to be compliant if they don't even know the rules, right?” Timestamps 00:00 Introduction to ClearTech Loop 00:26 Meet Maybelyn Plecic 01:29 Shadow AI: IT problem, security problem, or both? 01:54 Why shadow AI starts with trust 03:00 AI is moving faster than governance 04:47 AI generated content, visibility, and accountability 06:35 How language around AI is changing 08:43 Using AI to automate the boring work 10:40 How AI changes the CISO conversation 12:33 Non human identities and the importance of shared language 13:05 Workflow questions become security questions 14:26 Prompt injection, AI defaults, and training gaps 15:47 What AI defense means beyond tools 17:30 Why AI guidance has to match each team 18:45 Closing thoughts Guest Bio Maybelyn Plecic is the Manager of Training and Adoption at Network to Code. She specializes in helping teams make technical change practical, secure, and usable. Her work spans security posture, compliance initiatives, technical enablement, training strategy, and customer adoption. She brings a builder’s perspective to AI security, with a focus on making complex technology easier for people to understand and use responsibly. Additional Resources • Maybelyn Plecic website: https://www.maybelynplecic.com/ • Network to Code Resource Center: https://networktocode.com/resources/resource-center/ • NIST AI Risk Management Framework: https://www.nist.gov/itl/ai-risk-management-framework • Season 1 ClearTech Loop: https://www.buzzsprout.com/2248577 Follow ClearTech Loop for more conversations on AI security, cybersecurity leadership, AI governance, shadow AI, non human identities, and enterprise technology strategy. 🎧 Listen: In Buzzsprout Player ▶ Watch on YouTube: https://www.youtube.com/@ClearTechResearch/videos 📰 Subscribe to the Newsletter: https://www.linkedin.com/newsletters/7346174860760416256/

## Transcript

[0:07] SPEAKER_01: Hey y'all, thanks so much for joining.
[0:09] SPEAKER_01: I'm Joe Peterson.
[0:10] SPEAKER_01: I am the CIO of Clarify 360.
[0:13] SPEAKER_01: I'm the chief analyst for ClearTech Research.
[0:15] SPEAKER_01: And we're here today with ClearTech Loop.
[0:17] SPEAKER_01: We're on the move and in the know.
[0:19] SPEAKER_01: We're here with Maybelline Plechic, Manager of Training and Adoption at Network to Code.
[0:25] SPEAKER_01: Hi, Maybelline.
[0:26] SPEAKER_01: Hi.
[0:27] SPEAKER_00: Hi, everyone.
[0:28] SPEAKER_00: I'm uh definitely been a tech person for a long time.
[0:31] SPEAKER_00: Uh CISSP certified, AWS Cloud, many certifications.
[0:36] SPEAKER_00: But what matters to me, I feel like, is the experience and how you tell the story.
[0:39] SPEAKER_00: That's more important than anything.
[0:41] SPEAKER_00: Absolutely.
[0:42] SPEAKER_01: And thank you so much for joining today.
[0:44] SPEAKER_01: So, in case you're not following Maybelline, you should be.
[0:48] SPEAKER_01: She, as she mentioned, she is a CISSP and she specializes in helping enterprises strengthen their security posture.
[0:55] SPEAKER_01: She drives compliance initiatives and she navigates complex security audits.
[1:01] SPEAKER_01: Probably more fun than she should be having, really, truly.
[1:05] SPEAKER_01: She empowers teams and organizations to achieve their highest communication security, data protection, and regulatory compliance standards while eliminating waste and maximizing value delivery.
[1:18] SPEAKER_01: She's a distinguished engineer in technical enablement and a servant and a servant leader.
[1:24] SPEAKER_01: So I'm happy to have her here today.
[1:26] SPEAKER_01: Well, thank you for having me.
[1:27] SPEAKER_01: Excited.
[1:29] SPEAKER_01: And as you guys know, we ask three of the moment questions around AI security.
[1:35] SPEAKER_01: So I'm going to just start in right with those with Maybelline.
[1:38] SPEAKER_01: A first one, give me your thinking around shadow AI.
[1:42] SPEAKER_01: Is it an IT problem, a security problem, both, neither?
[1:47] SPEAKER_01: And second part of the question, how are CISOs and CIOs addressing shadow AI in their environments?
[1:54] SPEAKER_01: Awesome.
[1:54] SPEAKER_00: Well, thank you for that question.
[1:55] SPEAKER_00: And such a very vital question to be answered, especially on the leadership side, right?
[2:01] SPEAKER_00: I think it's an issue on both sides, organizationally, especially.
[2:04] SPEAKER_00: Many are always looking for the best outcome, right?
[2:09] SPEAKER_00: On the business side.
[2:10] SPEAKER_00: We can have the best technical requirements and meet it, deliver it.
[2:13] SPEAKER_00: But if we're not answering business questions, such as one, is it secure, private, do we have governance?
[2:19] SPEAKER_00: Are we able to sustain anything?
[2:21] SPEAKER_00: Uh, efficiency and effectiveness, that's not going to satisfy any business, right?
[2:26] SPEAKER_00: And shadow IT, it's it's really a partnership.
[2:29] SPEAKER_00: I believe that trust is where it needs to start.
[2:32] SPEAKER_00: There needs to be an openness to be curious, to have the collaborative conversations, not penalize people because they're trying.
[2:39] SPEAKER_00: Right now, they're very challenged on upskilling fast.
[2:42] SPEAKER_00: AI is developing fast, whether we choose it or not.
[2:46] SPEAKER_00: Uh, and it's too easy to access.
[2:48] SPEAKER_00: And I feel like that is a challenge in general for most because what system is going to be able to track how quickly it develops?
[2:56] SPEAKER_00: It's developing in what milliseconds?
[2:59] SPEAKER_00: We have versions between Cloud, between Chat GPT, Deep, you know, any other AI, copilot, you name it, they're moving faster than we can think or imagine for that matter.
[3:10] SPEAKER_00: How do you plan to mitigate uh and create protections when we don't think and move that fast?
[3:17] SPEAKER_00: Uh, and this is gonna take a lot of partnerships on POC, proof of concept.
[3:22] SPEAKER_00: I believe that there needs to be uh sandboxes that are allowed, like actual development places that you can protect and set guidelines and don't just make statements or general statements, right?
[3:34] SPEAKER_00: We need actual concrete guidance that we can follow.
[3:38] SPEAKER_00: We can over-document, but who implements it?
[3:40] SPEAKER_00: That's always a question, right?
[3:42] SPEAKER_00: I mean, we've gone through that route of audits and how many athletes comply?
[3:46] SPEAKER_00: Are we documenting just for the sake of documenting?
[3:49] SPEAKER_00: Yeah, athlete gonna do something about it and say, all right, because you've done this, you know, you created this behavior or this process, someone's gonna go behind and athlete audit that.
[4:00] SPEAKER_00: And even if we had AI do all that work, who's gonna go and audit it as a person?
[4:06] SPEAKER_00: We're all going to be uh responsible for a lot of data.
[4:10] SPEAKER_00: We're gonna be responsible for a lot of things happening fast.
[4:13] SPEAKER_00: And I feel like there's a dilemma and there's a gap in how fast we can decide based on the current information that we're being provided and even the firewalls that we built.
[4:23] SPEAKER_00: It's gonna have to evolve quickly to anticipate.
[4:26] SPEAKER_00: I can't imagine how malicious actors are taking advantage of this, right?
[4:30] SPEAKER_00: Between the deepfake videos out there, it's very disturbing that you can't even tell an AI video generated now.
[4:38] SPEAKER_00: There's no also tagging or signatures to figure out.
[4:43] SPEAKER_00: And we've had those conversations, Joe.
[4:45] SPEAKER_00: You know this.
[4:46] SPEAKER_01: I know that's why I'm laughing.
[4:48] SPEAKER_01: I put up a cat video.
[4:49] SPEAKER_01: You guys, if you follow me, you know I put up stupid cat videos with security.
[4:54] SPEAKER_01: And Maybelline and I connected over my stupid cat videos.
[4:58] SPEAKER_01: Like it's been a while now, but I put one out and I didn't know it was AI.
[5:03] unknown: Yeah.
[5:04] SPEAKER_01: And we got blowback.
[5:06] SPEAKER_01: Poor Maybelline said something about it, and she got blowback for you know contributing to AI slop.
[5:12] SPEAKER_01: And I'm like, oh my heavens, you know, I I I apologize to her because I didn't know it was AI.
[5:19] unknown: No.
[5:20] SPEAKER_00: It's very understandable because it's not recorded, right?
[5:23] SPEAKER_00: I mean, it's uh it's it's actually interesting to me because I'm also a blogger.
[5:27] SPEAKER_00: I write documentations, and typically you need to credit the image, whether it's free or not, right?
[5:33] SPEAKER_00: And I feel that if you're creating um AI video of anything or even image or content, you should credit and say, yeah, this is AI generated, right?
[5:42] SPEAKER_00: I think that's a pretty standard thing, but I think that there's innocently whether they intend it or not, but many people are not realizing how much of an implication it has, right?
[5:54] SPEAKER_00: That those videos or memes or images can create a narrative that they might not intend to, right?
[6:01] SPEAKER_00: And I feel like that's another education that we need to build.
[6:04] SPEAKER_00: Securities on the umbrella is very broad, right?
[6:07] SPEAKER_00: Especially AI.
[6:08] SPEAKER_00: And I feel that the concerns that we have, we can only solve to a certain point, but we also need policies in here.
[6:15] SPEAKER_00: It's it can't be just done on the software, on the organizational level, on the local level.
[6:21] SPEAKER_00: We need actual regulatory things that will actually ensure that we implement and follow the compliant regular regulations.
[6:30] SPEAKER_00: And if they don't understand AI, how are they going to build good policies?
[6:34] SPEAKER_00: That's my question.
[6:35] SPEAKER_01: That's that's that's a fair and you said a couple things that you know really resonated with me.
[6:41] SPEAKER_01: One, when before we even started recording, you said something to me, and I thought, oh my gosh, that's that's just our language is developing around AI.
[6:49] SPEAKER_01: And what I mean by that is 25 years ago, we nobody used the term googling, but now we use the term googling because it is we look up something, and you said that your husband says that you're always clotting.
[7:02] SPEAKER_01: Yeah, that is gonna become a term, you know that, right?
[7:06] SPEAKER_01: He he probably just he probably just coined a term, didn't know it.
[7:10] SPEAKER_01: You're clotting, yeah.
[7:11] SPEAKER_00: That is it, that is it.
[7:13] SPEAKER_00: And he and you know, I'm trying to control the time I spend and things just because it is very, very addictive, sadly, right?
[7:22] SPEAKER_00: Because you're seeing how well, I mean, imagine a workflow, something you've been trying to get done for years, and you do this in one weekend.
[7:31] SPEAKER_00: That's how efficient it is.
[7:33] SPEAKER_00: So, who will not be tempted to want to be effective, efficient, especially when KPIs for work where there's efficiency, effectiveness, right?
[7:43] SPEAKER_00: This is more accurate than a human being on documentation, unfortunately.
[7:46] SPEAKER_00: And there's no misspelling, right?
[7:50] SPEAKER_00: I've been looking at it for anyone as OCD about spelling.
[7:54] SPEAKER_00: You look at the screen, you're troubleshooting, and no misspelling.
[7:58] SPEAKER_01: I know it's it's crazy.
[8:00] SPEAKER_01: It is, it is, but you you also said something that gives me a little bit of hope because you you hit on something that so I'm I'm a glass half full girl, I'm an optimist, right?
[8:10] SPEAKER_01: And I get the feeling you are too.
[8:12] SPEAKER_01: Yes, I'm feeling that AI is an opportunity for the IT executives to, you know, because we tend to build silos, right?
[8:22] SPEAKER_01: The security folks sit over here and the IT folks sit over here.
[8:25] SPEAKER_01: And what's happening, I think, with AI is that these groups in companies where they didn't talk before, because there are groups that do talk, right?
[8:34] SPEAKER_01: But they're coming together because they have to to talk about this problem.
[8:37] SPEAKER_01: Because it takes both of their mindsets and and their skills to come and talk about this.
[8:43] SPEAKER_00: I agreed 100%.
[8:44] SPEAKER_00: If anything, even my my boss today, right?
[8:48] SPEAKER_00: It only took a month for me to build nine applications, right?
[8:52] SPEAKER_00: Very quickly, did not take that long, but I'll caveat it that the reason why I was able to do that is because I was always strategic to begin with, that I had an understanding of many systems that we integrate, right?
[9:05] SPEAKER_00: And I knew what the outcome specifically business-wise we were looking for.
[9:09] SPEAKER_00: Because just like any developer, you can code all you want, but if you don't know what you're coding for, what use cases you're matching, it's a useless application, right?
[9:18] SPEAKER_00: Uh, and then the part that I've learned so far, I feel like I'm a product owner now because I have a constant backlog now of feature requests, and all our meetings now have become feature request conversations, okay?
[9:31] SPEAKER_00: So I'm like prioritizing, okay.
[9:34] SPEAKER_00: What's next?
[9:35] SPEAKER_00: We have a separate channel just for automation requests, right?
[9:39] SPEAKER_00: And what it's been fun, if anything, if what I like you said, glass hat full.
[9:44] SPEAKER_00: I encourage everyone to have actual great conversations better now, I feel like, because the mundane stuff, the boring stuff that we never wanted to look at, AI is great at consuming large data.
[9:56] SPEAKER_00: That's what it's for, right?
[9:58] SPEAKER_00: It's great at picking up patterns that we didn't have time to even think about.
[10:03] SPEAKER_00: So I feel like we need to leverage AI for the boring mundane automation related stuff.
[10:08] SPEAKER_00: And this also, I feel like source of truth has been a problem.
[10:12] SPEAKER_00: Like Apple great data that's at have issues with garbage in and garbage out.
[10:18] SPEAKER_00: For sure.
[10:19] SPEAKER_00: But since we've done this, we're actually finding out all those gaps and fixing them, resolving them.
[10:26] SPEAKER_01: Like, oh we have to, we can't fix the candle down the road anymore.
[10:30] SPEAKER_01: Like, like, like, come on, for years we would just, you know, oh that data's over there, oh, that data's over here, and now it matters.
[10:40] SPEAKER_00: It matters more, even more so now.
[10:42] SPEAKER_00: Um, I feel that for CISO, I think they have a different, I think the conversations are transforming differently.
[10:50] SPEAKER_00: I feel like because of AI, now they have even more visibility.
[10:55] SPEAKER_00: Now they can ask more strategic questions.
[10:58] SPEAKER_00: Before it was very operational, just trying to breathe, address the fire.
[11:03] SPEAKER_00: Now you can kind of compartmentalize a little bit better, you can create a better roadmap because you're finding, oh, we're weak here, we have strengths here.
[11:12] SPEAKER_00: Now, how do we collaborate better with IT, with business folks, the board, our you know, go-to-market teams, our training or adoptions, anyone else, typical people that felt uncomfortable discussing their workflows, because we can conversate more in a business that they can understand, you're like, oh, I can see it now.
[11:34] SPEAKER_00: I mean, imagine when you're trying to get uh business requirements or product requirements, you can't imagine something that you're just talking out constantly, you know, just abstractly.
[11:45] SPEAKER_00: To me, it's better to do show and tell.
[11:48] SPEAKER_00: Now, if I can do a requirement very quickly based on what I capture, right out, show it to them, they can give you better constructive, continuous feedback.
[11:58] SPEAKER_01: Right, exactly.
[11:59] SPEAKER_01: And engineers are visual learners, let's just face it.
[12:02] SPEAKER_01: We are visual learners.
[12:04] SPEAKER_01: So let me ask you another question as a CISSP, because I I just am fascinated by NHIs.
[12:11] SPEAKER_01: I don't know about you, but I'm I'm just fascinated by them.
[12:14] SPEAKER_01: And I'm fascinated with like these super agents creating other agents and permissions.
[12:20] SPEAKER_01: And okay, as you think about NHIs and you think about enabling them, what what are a couple things that you think are super important to consider?
[12:33] SPEAKER_00: Uh I think language is important, right?
[12:36] SPEAKER_00: The terminologies that we use.
[12:38] SPEAKER_00: We can easily make misconstrue terminologies, whether technical or business.
[12:44] SPEAKER_00: Uh plain, simple language sometimes, basic terminologies on how we use it is more important than models or acronyms.
[12:52] SPEAKER_00: It drives me crazy when people mention acronyms and they're like, You're assuming everyone knows whether it's business or technical, right?
[12:59] SPEAKER_00: Yeah.
[13:00] SPEAKER_00: I feel like you have to have a common language, common perspective.
[13:04] SPEAKER_00: Yeah, that's good.
[13:05] SPEAKER_00: What's how is your day-to-day?
[13:07] SPEAKER_00: I mean, I've done adoption conversations in the past, and I really just ask simple things.
[13:11] SPEAKER_00: I was like, hey, you know, what is your typical day-to-day, right?
[13:15] SPEAKER_00: That you know, what do you do?
[13:16] SPEAKER_00: And then they just describe everything they know.
[13:19] SPEAKER_00: It's like, hey, I come in from work, check in, I check this, I set up my presentation, I need to connect this, I have to log in, get someone else to take care of my calendar.
[13:28] SPEAKER_00: Those little conversations, we take it for granted, but those are all potential security questions, access questions, efficiencies, troubleshooting, right?
[13:39] SPEAKER_00: Now, same case with AI, right?
[13:42] SPEAKER_00: Now it starts out with a friendly client, but simple things like not tracking the history, right?
[13:48] SPEAKER_00: Are you making sure you're not trading the AI?
[13:50] SPEAKER_00: I think that we need to do the basic walkthrough of onboarding anyone for that matter, right?
[13:57] SPEAKER_00: Because they can just take the default, and if you're not monitoring well, every provider that pushes AI, they could be technically capturing things they shouldn't, right?
[14:08] SPEAKER_00: Even I'm seeing in you know, prompt injection is another way from what I'm seeing that uh are being used to try to remind people on how not to pass information that they shouldn't.
[14:20] SPEAKER_00: But even then, it's a very delicate balance of are you inconveniencing someone to force them to do shadow AI, right?
[14:29] SPEAKER_00: Yeah, or are you gonna enable them from the start and say, hey, I understand everyone's curious, interested, and they want to all save time, whatever it is.
[14:40] SPEAKER_00: A lot of demos, lunch and learns need to happen.
[14:43] SPEAKER_00: You can't do e-learning necessarily because nobody has time to sit for whatever reason.
[14:49] SPEAKER_00: Yeah, okay if you get five minutes, even on a meeting, right?
[14:53] SPEAKER_00: Especially for C levels, you got them, they're out of your ear already if you didn't make it interesting.
[14:59] SPEAKER_00: I think now what we have to do better storytelling, visual storytelling, hands-on, a lot more hands-on, like live hands-on.
[15:10] SPEAKER_00: I I do that even for my onboarding for new employees that we have, and I say, Hey, here's a few slides, here's a quick link, all your links are available.
[15:19] SPEAKER_00: But guess what?
[15:19] SPEAKER_00: We're doing this hands-on together.
[15:23] SPEAKER_01: Because you get their attention, that's such a good point because you hold their attention in that moment, right?
[15:29] SPEAKER_01: And and you're imparting something that's really important, and I I like that.
[15:34] SPEAKER_01: That's an interesting way to kind of think about the problem.
[15:38] SPEAKER_01: Yeah, um, let me ask you our last question.
[15:41] SPEAKER_01: When you hear the term AI defense, what comes to mind for you?
[15:47] SPEAKER_00: I know by default, it's some type of software, right?
[15:50] SPEAKER_00: But I feel like it's more of mindset.
[15:54] SPEAKER_00: The AI defense to me is all the due diligence, the communication itself, right?
[16:01] SPEAKER_00: Are you up front with your policies?
[16:04] SPEAKER_00: Are you ensuring that they don't?
[16:07] SPEAKER_00: I mean, how do you expect someone to be compliant if they don't even know the rules?
[16:12] SPEAKER_00: Right.
[16:12] SPEAKER_00: And then, second, is it easy for people to follow?
[16:15] SPEAKER_00: If if the policy or instructions or process is so nuanced, there's too much of a hassle to follow, they're not gonna follow it at all.
[16:22] SPEAKER_00: No, especially our less technical people.
[16:26] SPEAKER_00: Correct.
[16:27] SPEAKER_00: Usually it's a very innocent thing.
[16:29] SPEAKER_00: Most people, it's not on purpose.
[16:31] SPEAKER_00: I feel like I would I try to give people the benefit of a doubt that most of them have good intent, right?
[16:37] SPEAKER_00: We have to start with that too.
[16:38] SPEAKER_00: We we cannot go in the route of judging people from the get-go and penalizing people.
[16:43] SPEAKER_00: If anything, we need to see how they work, just like that conversation I asked, I mentioned earlier.
[16:49] SPEAKER_00: How do they typically work?
[16:51] SPEAKER_00: Yeah, you need to see how they function, whether you have to shadow them just like shadow IT, watch what they do, right?
[16:59] SPEAKER_00: And see if there's little things that might be optimal for them that they didn't realize.
[17:04] SPEAKER_00: Uh, that's usually take.
[17:06] SPEAKER_00: I mean, how many people do you know even know how to use formulas for Excel or any of the special true?
[17:12] SPEAKER_01: I mean, right?
[17:13] SPEAKER_01: You and and and you bring up such an important point because you know, you know the marketing folks are all over AI, right?
[17:21] SPEAKER_01: But the accounting folks, maybe not so much.
[17:24] SPEAKER_01: So different people in the organization have different skill sets already as it relates to AI.
[17:30] SPEAKER_00: Yeah, and you bring up a great point that actually every person I talk to, I talk to in a different way to match their need, right?
[17:37] SPEAKER_00: Whether it's my operations person that I'm talking to that builds decks for our C levels, right?
[17:43] SPEAKER_00: They're building presentations, Google or PowerPoint.
[17:45] SPEAKER_00: That's a different conversation.
[17:47] SPEAKER_00: Customer success is a different conversation when it comes to customer health.
[17:51] SPEAKER_00: Talk to our sales or go to market is a different conversation.
[17:55] SPEAKER_00: And you have to keep it simple, right?
[17:57] SPEAKER_00: Very simple, just to what it is that they typically reach out for people that they could do themselves.
[18:03] SPEAKER_00: How do you self-serve?
[18:04] SPEAKER_00: And we need to do the same thing for shadow IT.
[18:07] SPEAKER_00: We need to create ways to make it self-serve so that IT is not bogged down with so many tickets, and then they have to deal with all the audits and security flags that they might have to deal with that could have been prevented if you decided from the get-go that you're embracing it.
[18:21] SPEAKER_00: So I feel like it's a top-down decision to make.
[18:25] SPEAKER_00: I feel that whoever's at the top, every leadership role from the top down has to make a conscious decision to align on what that messaging is supposed to be, so that everyone else can follow.
[18:38] SPEAKER_00: If you start from the bottom up, it's not gonna, it's gonna defeat it, right?
[18:43] SPEAKER_00: Be feasible in the long term.
[18:45] SPEAKER_00: Right.
[18:46] SPEAKER_01: Well, such a lovely conversation with you.
[18:49] SPEAKER_01: Thank you for taking the time to visit.
[18:51] SPEAKER_01: Have to have you come back.
[18:52] SPEAKER_01: Yes.
[18:53] SPEAKER_01: Um, I can't wait to hear what you think about MCP servers.
[18:56] SPEAKER_01: So that's that's the conversation for another day.
[19:00] SPEAKER_01: Yes.
[19:00] SPEAKER_01: You've been lovely, guys.
[19:02] SPEAKER_01: Thank you so much for joining, and we'll catch you next time.
[19:04] SPEAKER_00: Awesome.
[19:05] SPEAKER_00: Take care.

Shows are listed from Apple Podcasts' charts and directory, and Podcast Index where it is used. Every show and episode belongs to its publisher and plays from the publisher's own feed; this site hosts none of them.